Is it possible? Frequency division with op amps?

[ali6x944]

Hello everybody,
Is it possible to "Frequency division" with op amps, i know its working with d and jk flip flops but will it with op amps?

[T3sl4co1l]

Yes.  Make an injection-locked oscillator.

It will only work over a small range of input frequencies.

To truly recreate a D or JK or T f/f, you'd need to recreate the logic that make them up, inventing your own system of digital voltage thresholds, and gate circuits.  It would be extremely slow, also.

Tim

[GK]

This took about 5 minutes to conceptualize. A better (faster) version would use feedback to stop the op-amps from saturating and work with a reduced output voltage swing. A rather pointless circuit though and now I am off to bed.

LTspice file: https://www.eevblog.com/forum/beginners/is-it-possible-frequency-division-with-op-amps/?action=dlattach;attach=199886

[vk6zgo]

Yes.  Make an injection-locked oscillator.

It will only work over a small range of input frequencies.

To truly recreate a D or JK or T f/f, you'd need to recreate the logic that make them up, inventing your own system of digital voltage thresholds, and gate circuits.  It would be extremely slow, also.

Tim

Back in the Dreamtime,when bistable circuits were made using 6SN7GTs,TV Sync Pulse Generators for OB use could not use cascaded "divide by two"circuits as Studio SPGs did,due to size limitations,so they used,"injection locked" free running multivibrators.

The advantage of this approach was that you could obtain large division ratios,thus minimising component count.
The disadvantage was that they were "touchy" to adjust.

The old Marconi one at Perth Tech College had a small CRT,which showed a specific number of dots when everything was set up correctly.
It should be possible to create a far more stable device with modern Op Amps.

[IconicPCB]

UJTs come to mind...not as modern as opamps..

[Zero999]

The answer is yes but why would you want to?

Comparators are more suited to this than op-amps so this may make sense if you're using an LM339 and have spare comparators on the ID.

[FenixEletronica]

@Zero999 I am learning how OpAmps work and would like to ask you to analyze the following circuit, which seems to me as an OpAmp frequency divider:

The first, left hand OpAmp is getting a 12 V signal @pin 3...
The 5 Hz singal drives the "balance LEDs" as the measurement is being done. This signal feeds a logic gate to drive panel LEDs and also feeds the control logic...

My issue is that all components measurements looks like okay, while V14 quickly heats uo to 55ºC - I shut it down before any smoke shows up.
 

[Zero999]

@Zero999 I am learning how OpAmps work and would like to ask you to analyze the following circuit, which seems to me as an OpAmp frequency divider:

The first, left hand OpAmp is getting a 12 V signal @pin 3...
The 5 Hz singal drives the "balance LEDs" as the measurement is being done. This signal feeds a logic gate to drive panel LEDs and also feeds the control logic...

My issue is that all components measurements looks like okay, while V14 quickly heats uo to 55ºC - I shut it down before any smoke shows up.
 

What schematic? It appears you either forgot to upload it, or there was a software glitch.

Click Attachments and other options to upload an image of the schematic.

[KE5FX]

You could build a regenerative divider with an opamp and a diode (or other) mixer, I suppose.  It should yield a cleaner output waveform than a multivibrator, at least if you can incorporate some sort of AGC mechanism.

[Terry Bites]

Yeah, but why. It isn't a real world opamp application. The only person who tried it died in a santorium.

[mawyatt]

A passive mixer, altho not an op-amp, can be configured as a divide by 2 frequency divider, and requires no power supply.

Best,

[KE5FX]

x2 multiplier, sure.  But how would a divider work?

[T3sl4co1l]

There is the active case, where the mixer is wired as a squaring function, put in a feedback loop to obtain a sqrt(x) function, and a flip-flop added to solve for the (i.e. to alternate it) which is otherwise ambiguous from the arithmetic sqrt; or perhaps the arguably-less-passive case with a parametric amp, powered by the input and tuned to f_in/2, but I'm not sure that that was intended here.

Tim

[KE5FX]

Right, I can't think of a way to build a divider with just a passive mixer.  To build a divider, something has to retain some state...

[mawyatt]

x2 multiplier, sure.  But how would a divider work?

Maybe this can help!

For a Mixer call 3 connections Fi (Input Port), Fx and Fy. Connect Fy to Fx with Low Pass Filter, and supply signal to Fi.

Assuming ideal multiplication Signal Frequency at Fx is Fi +- Fy, and Fy is Fi -Fy due to LPF.

Thus Fy = Fi - Fy, so 2Fy = Fi

Fy = Fi/2

Edit: Here's a quick sketch.

Best,

[KE5FX]

Sure, that's a regenerative divider.  But there's no way it'll work without gain.

[FenixEletronica]

@Zero999 I remembered to attach a .png file, but didn't check after posting.
My bad!

[amyk]

https://en.wikipedia.org/wiki/Frequency_divider#Analog

[MarkT]

Hello everybody,
Is it possible to "Frequency division" with op amps, i know its working with d and jk flip flops but will it with op amps?

There might be a novelty way using an opamp with phase-reversal.  Maybe.  In general phase reversal is avoided but here it might be leveraged to add state?
You can use an opamp as part of a PLL of course, or as an oscillator, so more conventional approaches are doable.

In general though possible != practical.  Chocolate teapots and gallium teaspoons are possible...  Best to pick the right tool for the job usually!

[Xena E]

Too lazy to read the whole thread.

Has anyone mentioned squaring of the input pulse and feeding a staircase generator?

Can be cascaded for greater division by n ratios.

[Zero999]

@Zero999 I remembered to attach a .png file, but didn't check after posting.
My bad!

That's just a band stop filter. I don't see how it can divide the frequency. The RC network (R1 to R3 and C1 to C3) blocks 5Hz and allows the other frequencies though, which are amplified by the op-amp, which has a gain of 1+22/18 = 2.22.